WO2006052508A2 - An integrated image processor - Google Patents

An integrated image processor Download PDF

Info

Publication number
WO2006052508A2
WO2006052508A2 PCT/US2005/039285 US2005039285W WO2006052508A2 WO 2006052508 A2 WO2006052508 A2 WO 2006052508A2 US 2005039285 W US2005039285 W US 2005039285W WO 2006052508 A2 WO2006052508 A2 WO 2006052508A2
Authority
WO
WIPO (PCT)
Prior art keywords
processor
substrate
pixel
integrated image
depth
Prior art date
Application number
PCT/US2005/039285
Other languages
English (en)
French (fr)
Other versions
WO2006052508A3 (en
Inventor
John Iseln Woodfill
Ronald John Buck
Gaile Gibson Gordon
David Walter Jurasek
Terrence Lee Brown
Original Assignee
Tyzx, Inc.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tyzx, Inc. filed Critical Tyzx, Inc.
Priority to JP2007540360A priority Critical patent/JP5306652B2/ja
Priority to EP05821343A priority patent/EP1810220A4/en
Publication of WO2006052508A2 publication Critical patent/WO2006052508A2/en
Publication of WO2006052508A3 publication Critical patent/WO2006052508A3/en

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/94—Hardware or software architectures specially adapted for image or video understanding
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/50—Depth or shape recovery
    • G06T7/55—Depth or shape recovery from multiple images
    • G06T7/593—Depth or shape recovery from multiple images from stereo images
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/94—Hardware or software architectures specially adapted for image or video understanding
    • G06V10/955—Hardware or software architectures specially adapted for image or video understanding using specific electronic processors
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle

Definitions

  • the processing of these images is computationally intensive and requires the handling of large amounts of data.
  • the processing has substantial latency because it involves a series of calculations where one type of calculation is performed over an entire image before moving on to a next type calculation.
  • the output arrays from these calculations are usually large, sometimes including for each position in the image, depth information, intensity information, and color information. It would be useful to reduce the output arrays to ease calculation requirements for subsequent processing. It would also be useful to enable real-time image processing by reducing the latency.
  • Figure 1 is a block diagram illustrating an embodiment of an integrated image processor.
  • Figure 2 is a block diagram illustrating an embodiment of an integrated image processor.
  • Figure 3 is a block diagram illustrating an embodiment of an integrated image processor.
  • Figure 4 is a block diagram illustrating an embodiment of a pixel handling processor.
  • the invention can be implemented in numerous ways, including as a process, an apparatus, a system, a composition of matter, a computer readable medium such as a computer readable storage medium or a computer network wherein program instructions are sent over optical or electronic communication links.
  • these implementations, or any other form that the invention may take, may be referred to as techniques.
  • a component such as a processor or a memory described as being configured to perform a task includes both a general component that is temporarily configured to perform the task at a given time or a specific component that is manufactured to perform the task.
  • the order of the steps of disclosed processes may be altered within the scope of the invention.
  • An input interface is configured to receive pixel data from two or more images.
  • a pixel handling processor disposed on the substrate is configured to convert the pixel data into depth and intensity pixel data.
  • a foreground detector disposed on the substrate is configured to classify pixels as foreground or not foreground.
  • a projection generator disposed on the substrate is configured to generate a projection of the depth and intensity pixel data.
  • a segmentation processor disposed on the substrate is configured to generate a list of objects and their descriptions including height, extent, number, and/or color, hi various embodiments, an application processor disposed on the substrate is configured to use results of the other processors to produced outputs for user applications such as count people, count cars, count objects, detect objects, track objects, identify obstacles, detect navigable paths for vehicles, and or identify tailgaters through a door.
  • the source of the images does not change location between images and an object of interest does not change locations between images; the source of the images does not change location between images and an object of interest does change locations between images; the source of the images does change location between images and an object of interest does not change locations between images; and the source of the images does change locations between images and an object of interest does change locations between images.
  • FIG. 1 is a block diagram illustrating an embodiment of an integrated image processor.
  • integrated image processor 100 includes input interface 102, pixel handling processor 104, and foreground detector processor 106.
  • Input interface 102 inputs include pixel data from two or more sources.
  • the sources provide image data that allow depth information to be calculated
  • the pixel data include intensity information, color information (for example, red, green, blue, luminance, other spectral bands such as UV or thermal IR, chrominance, saturation, and/or intensity information), vertical synchronization information, and horizontal synchronization information
  • there are two sources, stereo sources, providing image data hi various embodiments, each source produces an image of 512 x 512 pixels occurring 20 or 200 times per second, hi various embodiments, source data size is 512 x 2048 with 8-bit or 10-bit pixels, hi various embodiments, the separation between two source cameras is 3 cm, 5 cm, 22 cm, or 33 cm.
  • the field of view is 45 degrees or 50 degrees based at least in part on the lens of the image source.
  • Input interface 102 outputs pixel data to pixel handling processor 104.
  • Pixel handling processor 104 output(s) include one or more of the following: depth information, disparity information, data validity information, and image characteristics such as average intensity.
  • Pixel handling processor 104 grabs the one or more images of pixel data, rectifies the one or more images by removing distortions from lenses and cameras based on calibration parameters, and calculates depth or disparity.
  • Foreground detector processor 106 inputs include depth information or disparity information, and/or data validity information.
  • Foreground detector processor 106 outputs include an indication of which pixels are part of the foreground and which pixels are not part of the foreground.
  • foreground detection comprises modeling the background. Building a model of the background and updating that model over time allows the classification of pixels into foreground or background.
  • the background pixels are the pixels that remain relatively stable over a relatively long period of time.
  • Foreground pixels are pixels corresponding to objects that do not remain relatively stable.
  • a model of the background is constructed by identifying pixels that remain constant to within a certain variation in range or disparity.
  • a model of the background is constructed by identifying pixels that remain constant to within a certain variation in color or intensity and/or range over a period of time (for example, a time based at least in part on a relevant time constant for the motion of interest in the source images) and indicating that these pixels are background pixels.
  • input interface 102, pixel handling processor 104, and foreground detector processor 106 are arranged in a pipeline architecture where each block is processing in parallel or at the same time. In some embodiments, only a subset of image information is required by each block so that, depending on the latency of each block and the size of the subset of information required by each block, a number of blocks are operating on pixel data from the same frame period at the same time, where a frame period is the time between frames of image data, hi various embodiments, input interface 102, pixel handling processor 104, and foreground detector processor 106 are integrated circuits disposed on a common substrate or as part of a common integrated circuit package.
  • FIG. 2 is a block diagram illustrating an embodiment of an integrated image processor.
  • integrated image processor 200 includes input interface 202, pixel handling processor 204, and projection generator 206.
  • Input interface 202 inputs include pixel data from two or more sources. The sources provide image data that allow depth information to be calculated.
  • the pixel data include intensity information, color information (for example, red, green, blue, luminance, chrominance, saturation, and/or intensity information), vertical synchronization information, and horizontal synchronization information.
  • Input interface 202 outputs pixel data to pixel handling processor 204.
  • Pixel handling processor 104 output include one or more of the following: depth information, disparity information, and validity information.
  • Pixel handling processor 204 grabs the one or more images of pixel data, rectifies the one or more images by removing distortions from lenses and cameras, and calculates depth or disparity.
  • Projection generator 206 inputs include depth information or disparity information, an optional input mask, and/or validity information.
  • Projection generator 206 outputs include one or more of the following: a 3 -dimensional (3D) data set or a 2-dimensional projection of the 3D data.
  • the 3-dimensional data set is a transformed, quantized 3D representation of the input data. The quantization is independent for each axis.
  • Input depth pixels are mapped into cells of the 3D map based on desired quantization and camera calibration parameters.
  • Statistics for pixels falling into each cell may be maintained including for example: count, minimum and/or maximum value on a given axis, and/or color characteristics, hi some embodiments, a 3D rigid transform, consisting of a 3D rotation, translation, and scaling, is applied during the mapping into the 3D projection volume.
  • the 2-dimensional data set is an embodiment of the 3D data set in which one dimension has quantization 1, i.e. is the projection of the 3D data along that axis.
  • useful statistics for example, minimum spatial extent, intensity, or color values, maximum spatial extent, intensity, or color values, histograms are stored as part of the projection data set.
  • a top down projection of a room it is useful (for example, for locating the floor or a person) to know the minimum or maximum height for a given location; this can be derived by storing the lowest or highest spatial location mapped to a cell.
  • the projection statistics are rescaled. For example, a closer object to the imaging source of a given size will have a greater number of pixels representing it as compared to a farther object of the same given size.
  • a correction to rescale the pixel counts for a given projection cell based on the distance to the image source can be calculated and applied for the 2-dimensional or 3-dimensional projection data set.
  • a threshold is applied before rescaling, after rescaling, or before and after rescaling.
  • input interface 202, pixel handling processor 204, and projection generator 206 are arranged in a pipeline architecture where each block is processing in parallel or at the same time, hi some embodiments, only a subset of image information is required by each block so that, depending on the latency of each block and the size of the subset of information required by each block, a number of blocks are operating on pixel data from the same frame period at the same time.
  • input interface 202, pixel handling processor 204, and projection generator 206 are integrated circuits disposed on a common substrate or as part of a common integrated circuit package.
  • FIG. 3 is a block diagram illustrating an embodiment of an integrated image processor.
  • integrated image processor 300 includes input interface 302, pixel handling processor 304, foreground detector processor 306, projection generator 308, segmentation processor 310, and application processor 312.
  • Input interface 302 inputs include pixel data from two or more sources. The sources provide image data that allow depth information to be calculated.
  • the pixel data include intensity information, color information (for example, red, green, blue, luminance, chrominance, saturation, other spectral bands such as UV or thermal IR, and/or intensity information), vertical synchronization information, and horizontal synchronization information.
  • Input interface 302 outputs pixel data to pixel handling processor 304.
  • Pixel handling processor 104 outputs include two or more of the following: depth information, disparity information, and validity information.
  • Pixel handling processor 304 accepts a coordinate stream of pixel data from two or more sources, rectifies the pixel streams by removing distortions from lenses and cameras, and calculates depth or disparity.
  • Foreground detector processor 306 inputs include depth information or disparity information, and/or validity information.
  • Foreground detector processor 306 outputs include an indication of which pixels are part of the foreground and which pixels are not part of the foreground. The indication of which pixels are part of the foreground is indicated by a data array of one bit per pixel.
  • foreground detection comprises modeling the background. The classification of pixels into foreground or background comes from building a model of the background and updating that model over time. The background pixels are the pixels that remain relatively stable over a relatively long period of time.
  • Projection generator 308 inputs include depth information or disparity information, and/or validity information.
  • Projection generator 308 inputs include depth information or disparity information, an optional input mask, and/or validity information.
  • Projection generator 308 outputs include one or more of the following: a 3-dimensional data set or a 2-dimensional projection of the 3D data.
  • Segmentation processor 310 inputs include the outputs of prpj ection generator 308.
  • Segmentation processor 310 calculations include smoothing, thresholding, and connecting elements of the 2-dimensional projection data set or 3- dimensional data set to create a list of objects and descriptors of those objects. For example, in a 2-dimensional projection data set parallel to the floor of a room, objects are segmented using a smoothing filter on the counts of neighboring cells, applying a threshold to the smoothed values to help in detecting object boundaries, and identifying locally connected regions remaining as objects of interest. Each object can be described in terms of spatial location 3D physical extent, color, etc. In some embodiments, one or more edge detection filters are used to identify boundaries around objects. In some embodiments, segmentation processor 310 detects which areas in the top or plan view projection represent an object (for example, a person).
  • Application processor 312 inputs include the outputs of the segmentation processor 310.
  • application processor 312 can count objects, classify objects, track objects, or detect patterns in the objects. For example, application 312 can count cars by identifying objects with an extent that is large enough to be a vehicle on top of a road surface and counting each object as a car. hi another example, application 312 can classify the vehicles to differentiate between cars and trucks using the different descriptors of the objects - cars are not as tall as, are narrower than, and are less long than a truck. As another example, in situations where a card key swipe is required to enter through a door, identifying the pattern of two people moving through the door on one card swipe can indicate that there is breach of security. In some embodiments, application processor 312 tracks people by comparing the location of the people in a current frame to prior locations and assigns person identifiers based on the spatial correspondence of the people in the frames.
  • input interface 302, pixel handling processor 304, foreground detector processor 306, projection generator 308, segmentation processor 310, and application processor 312 are arranged in a pipeline architecture where each block is processing in parallel or at the same time. In some embodiments, only a subset of image information is required by each block so that, depending on the latency of each block and the size of the subset of information required by each block, a number of blocks are operating on pixel data from the same frame period at the same time, hi various embodiments, input interface 302, pixel handling processor 304, foreground detector processor 306, projection generator 308, segmentation processor 310, and application processor 312 are integrated circuits disposed on a common substrate or as part of a common integrated circuit package.
  • FIG. 4 is a block diagram illustrating an embodiment of a pixel handling processor.
  • pixel handling processor 400 in Figure 4 is used to implement 104 of Figure 1, 204 of Figure 2, and 304 of Figure 3.
  • pixel handling processor 400 includes image grabbing processor 402, rectify processor 404, and depth/disparity processor 406.
  • the pixel handling processor does not include a rectify processor.
  • Image grabbing processor 402 inputs include image information from two or more sources in a roughly synchronous manner.
  • Image information depends on the source and can include intensity and color information in a number of formats (for example, RGB - red, green, blue, HSV - hue, saturation, value, HSB - hue, saturation, brightness, or YUV - luminance and two chrominance values, or any other appropriate image information format including other spectral bands) for pixels (image elements) and vertical and horizontal synchronization information.
  • Image grabbing processor 402 outputs synchronous image information for two or more images in a single format.
  • Rectify processor 404 has as input the outputs from image grabber 402 and source (for example, cameras, color imagers, monochrome imagers, etc.) predetermined calibration parameters (for example, relative imager positions and orientations, lens principal points, focal lengths, and lens distortion parameters). Rectify processor 404 "rectifies", corrects for geometric misalignment and lens distortion, each input source (in some cases removal of all distortions and misalignments is not possible). Lens and camera distortions and position corrections are removed by rotation, translation, magnification, and distortion removal operations to the images. In some embodiments, rotation, translation, and magnification are achieved using matrix operations on the image information, hi some embodiments, distortion removal is achieved using a polynomial distortion removal model.
  • mappings from rectified image pixel coordinates to distorted image pixel coordinates can be stored in the form of precalculated lookup tables. In some embodiments, the mappings can be computed on the fly during rectification of the pixel streams. In some embodiments, the mappings do not match any exact distorted image pixel coordinates so that interpolation is required to derive the rectified pixel coordinate and/or pixel value. In various embodiments, interpolation is linear interpolation, spline interpolation, or any other appropriate interpolation. Rectify processor 404 operations require a subset of image information for calculation of each pixel. In some embodiments, a pipelining architecture is used to perform calculations in parallel where only the subset of image information required for calculating a given pixel information is required by rectify processor 404.
  • Depth/disparity processor 406 inputs include the output information from rectify processor 404.
  • Depth/disparity processor 406 calculates the disparity of pixels in the input images by correlating pixel data between the two or more images and/or the depth, or range from the source, based at least in part on calibration parameters. The depth and/or disparity calculation is performed by correlating one or more pixels along a dimension of the image.
  • an array of pixel values from one image is correlated with pixel values from another image, hi some embodiments, a non- parametric correlation is used, hi various embodiments, SLOG (sign bits of Laplacian of Gaussian), SAD (sum of absolute differences), SSD (sum of squared differences), etc., may be used for correlation.
  • validity information is provided as an output for the depth/disparity processor block.
  • Validity information can be derived from checking the correspondence between calculating the disparity between, for example, image 1 and image 2 and then image 2 and image 1. In some embodiments, if the pixel area in the image is smooth, then the depth information is less likely to be valid, hi various embodiments, intensity and/or color information is used to calculate depth and/or disparity.
  • image grabbing processor 402 rectify processor
  • image grabbing processor 402, rectify processor 404, and depth/disparity processor 406 are integrated circuits disposed on a common substrate or as part of a common integrated circuit package.
  • pixel handling processor 400 does not include depth/disparity processor 406.
  • pixel handling processor 400 has as input one or more images of information containing range, or depth, information for each pixel.
  • Foreground detector processor 106 or projection processor 206 can then operate on input range information based at least in part on the one or more input images with range information.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Software Systems (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Processing (AREA)
  • Image Analysis (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Optical Radar Systems And Details Thereof (AREA)
PCT/US2005/039285 2004-11-03 2005-10-31 An integrated image processor WO2006052508A2 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
JP2007540360A JP5306652B2 (ja) 2004-11-03 2005-10-31 集積された画像プロセッサ
EP05821343A EP1810220A4 (en) 2004-11-03 2005-10-31 INTEGRATED IMAGE PROCESSOR

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US62495404P 2004-11-03 2004-11-03
US60/624,954 2004-11-03

Publications (2)

Publication Number Publication Date
WO2006052508A2 true WO2006052508A2 (en) 2006-05-18
WO2006052508A3 WO2006052508A3 (en) 2007-04-26

Family

ID=36336962

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2005/039285 WO2006052508A2 (en) 2004-11-03 2005-10-31 An integrated image processor

Country Status (4)

Country Link
US (2) US7664315B2 (enrdf_load_stackoverflow)
EP (1) EP1810220A4 (enrdf_load_stackoverflow)
JP (1) JP5306652B2 (enrdf_load_stackoverflow)
WO (1) WO2006052508A2 (enrdf_load_stackoverflow)

Families Citing this family (28)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10361802B1 (en) 1999-02-01 2019-07-23 Blanding Hovenweep, Llc Adaptive pattern recognition based control system and method
US8352400B2 (en) 1991-12-23 2013-01-08 Hoffberg Steven M Adaptive pattern recognition based controller apparatus and method and human-factored interface therefore
US7904187B2 (en) 1999-02-01 2011-03-08 Hoffberg Steven M Internet appliance system and method
US7146260B2 (en) 2001-04-24 2006-12-05 Medius, Inc. Method and apparatus for dynamic configuration of multiprocessor system
US10298735B2 (en) 2001-04-24 2019-05-21 Northwater Intellectual Property Fund L.P. 2 Method and apparatus for dynamic configuration of a multiprocessor health data system
US7178049B2 (en) 2002-04-24 2007-02-13 Medius, Inc. Method for multi-tasking multiple Java virtual machines in a secure environment
WO2006052508A2 (en) * 2004-11-03 2006-05-18 Tyzx, Inc. An integrated image processor
US7337650B1 (en) 2004-11-09 2008-03-04 Medius Inc. System and method for aligning sensors on a vehicle
WO2009016614A2 (en) * 2007-08-02 2009-02-05 Emza Visual Sense Ltd. Universal counting and measurement system
US10425595B2 (en) * 2007-11-28 2019-09-24 Flir Systems, Inc. Modular camera systems and methods
US9358924B1 (en) 2009-05-08 2016-06-07 Eagle Harbor Holdings, Llc System and method for modeling advanced automotive safety systems
US8417490B1 (en) 2009-05-11 2013-04-09 Eagle Harbor Holdings, Llc System and method for the configuration of an automotive vehicle with modeled sensors
US8941726B2 (en) * 2009-12-10 2015-01-27 Mitsubishi Electric Research Laboratories, Inc. Method and system for segmenting moving objects from images using foreground extraction
KR101429250B1 (ko) * 2010-08-20 2014-09-25 주식회사 팬택 단계별 객체 정보 제공이 가능한 단말 장치 및 방법
US10200671B2 (en) * 2010-12-27 2019-02-05 3Dmedia Corporation Primary and auxiliary image capture devices for image processing and related methods
US9501834B2 (en) * 2011-08-18 2016-11-22 Qualcomm Technologies, Inc. Image capture for later refocusing or focus-manipulation
US9604581B2 (en) 2011-11-07 2017-03-28 Magna Electronics Inc. Vehicle vision system with color correction
CN103164851B (zh) * 2011-12-09 2016-04-20 株式会社理光 道路分割物检测方法和装置
US8886392B1 (en) 2011-12-21 2014-11-11 Intellectual Ventures Fund 79 Llc Methods, devices, and mediums associated with managing vehicle maintenance activities
US9269155B2 (en) * 2012-04-05 2016-02-23 Mediatek Singapore Pte. Ltd. Region growing method for depth map/color image
US10237528B2 (en) * 2013-03-14 2019-03-19 Qualcomm Incorporated System and method for real time 2D to 3D conversion of a video in a digital camera
US10326969B2 (en) 2013-08-12 2019-06-18 Magna Electronics Inc. Vehicle vision system with reduction of temporal noise in images
US9639951B2 (en) * 2014-10-23 2017-05-02 Khalifa University of Science, Technology & Research Object detection and tracking using depth data
US9639166B2 (en) * 2015-03-11 2017-05-02 Microsoft Technology Licensing, Llc Background model for user recognition
US9536287B1 (en) * 2015-07-09 2017-01-03 Intel Corporation Accelerated lens distortion correction with near-continuous warping optimization
US10282623B1 (en) * 2015-09-25 2019-05-07 Apple Inc. Depth perception sensor data processing
US10430994B1 (en) 2016-11-07 2019-10-01 Henry Harlyn Baker Techniques for determining a three-dimensional textured representation of a surface of an object from a set of images with varying formats
JP7021036B2 (ja) * 2018-09-18 2022-02-16 株式会社東芝 電子機器及び通知方法

Family Cites Families (47)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB1406511A (en) * 1973-01-19 1975-09-17 Ici Ltd Cellular polyurethanes
US3901595A (en) * 1974-02-13 1975-08-26 Bendix Corp Parallel line scanning system for stereomapping
US4574394A (en) * 1981-06-01 1986-03-04 Environmental Research Institute Of Mi Pipeline processor
US4524455A (en) * 1981-06-01 1985-06-18 Environmental Research Inst. Of Michigan Pipeline processor
US5054097A (en) * 1988-11-23 1991-10-01 Schlumberger Technologies, Inc. Methods and apparatus for alignment of images
US5592599A (en) * 1991-12-18 1997-01-07 Ampex Corporation Video special effects system with graphical operator interface
US5289577A (en) * 1992-06-04 1994-02-22 International Business Machines Incorporated Process-pipeline architecture for image/video processing
JP3679426B2 (ja) * 1993-03-15 2005-08-03 マサチューセッツ・インスティチュート・オブ・テクノロジー 画像データを符号化して夫々がコヒーレントな動きの領域を表わす複数の層とそれら層に付随する動きパラメータとにするシステム
US5531227A (en) * 1994-01-28 1996-07-02 Schneider Medical Technologies, Inc. Imaging device and method
WO1995020343A1 (en) * 1994-01-28 1995-08-03 Schneider Medical Technologies, Inc. Imaging device and method
US5559335A (en) * 1994-12-28 1996-09-24 The University Of Utah Rotating and warping projector/backprojector for converging-beam geometries
US5812787A (en) * 1995-06-30 1998-09-22 Intel Corporation Video coding scheme with foreground/background separation
GB9513515D0 (en) * 1995-07-03 1995-09-06 Sgs Thomson Microelectronics Expansion of data
JP3710548B2 (ja) * 1996-05-08 2005-10-26 松下電器産業株式会社 車両検出装置
US5982378A (en) * 1996-08-02 1999-11-09 Spatial Technology Inc. System and method for modeling a three dimensional object
US5870085A (en) * 1997-02-03 1999-02-09 Ati International Generating text strings
US6215898B1 (en) * 1997-04-15 2001-04-10 Interval Research Corporation Data processing system and method
US6064760A (en) * 1997-05-14 2000-05-16 The United States Corps Of Engineers As Represented By The Secretary Of The Army Method for rigorous reshaping of stereo imagery with digital photogrammetric workstation
DE19817720A1 (de) * 1998-04-21 1999-10-28 Heidelberger Druckmasch Ag Verfahren zur Bilddatenkomprimierung für Zwei-Farben-Bilder
JP3544524B2 (ja) * 1998-06-25 2004-07-21 松下電器産業株式会社 画像処理装置
US6594385B2 (en) * 1998-10-27 2003-07-15 Hewlett-Packard Development Company, L.P. Image compression of background and text tiles
WO2000034919A1 (en) * 1998-12-04 2000-06-15 Interval Research Corporation Background estimation and segmentation based on range and color
DE29912572U1 (de) * 1999-07-19 1999-09-16 Marantec Antriebs- und Steuerungstechnik GmbH & Co. KG, 33428 Marienfeld Vorrichtung zum Öffnen bzw. Schließen eines Tores
US6556704B1 (en) * 1999-08-25 2003-04-29 Eastman Kodak Company Method for forming a depth image from digital image data
DE60114498T2 (de) * 2000-04-11 2006-08-03 Hoya Corp. Darstellung der Leistung eines Brillenglases durch eine Indexanzeige in visuell verständlicher Weise
US7386170B2 (en) * 2000-06-30 2008-06-10 Texas Instruments Incorporated Image object ranking
US6788809B1 (en) * 2000-06-30 2004-09-07 Intel Corporation System and method for gesture recognition in three dimensions using stereo imaging and color vision
US6567099B1 (en) * 2000-11-15 2003-05-20 Sony Corporation Method and system for dynamically allocating a frame buffer for efficient anti-aliasing
US6792140B2 (en) * 2001-04-26 2004-09-14 Mitsubish Electric Research Laboratories, Inc. Image-based 3D digitizer
US20030007703A1 (en) * 2001-07-03 2003-01-09 Roylance Eugene A. Configurable image processing logic for use in image processing devices
JP2003109128A (ja) * 2001-07-25 2003-04-11 Fuji Electric Co Ltd 侵入者検出装置
US7003136B1 (en) * 2002-04-26 2006-02-21 Hewlett-Packard Development Company, L.P. Plan-view projections of depth image data for object tracking
US7227893B1 (en) * 2002-08-22 2007-06-05 Xlabs Holdings, Llc Application-specific object-based segmentation and recognition system
US7567274B2 (en) * 2002-12-09 2009-07-28 Frank Edughom Ekpar Method and apparatus for creating interactive virtual tours
KR100503820B1 (ko) 2003-01-30 2005-07-27 학교법인 포항공과대학교 시스톨릭 어레이를 이용한 멀티레이어 실시간 입체 영상정합 시스템 및 방법
US7224830B2 (en) * 2003-02-04 2007-05-29 Intel Corporation Gesture detection from digital video images
US7017492B2 (en) * 2003-03-10 2006-03-28 Quad/Tech, Inc. Coordinating the functioning of a color control system and a defect detection system for a printing press
GB2400287A (en) * 2003-04-02 2004-10-06 Autodesk Canada Inc Three-Dimensional Image Compositing
US7298898B2 (en) * 2003-04-15 2007-11-20 Honeywell International Inc. System and method for image segmentation
US7643025B2 (en) * 2003-09-30 2010-01-05 Eric Belk Lange Method and apparatus for applying stereoscopic imagery to three-dimensionally defined substrates
US7747067B2 (en) * 2003-10-08 2010-06-29 Purdue Research Foundation System and method for three dimensional modeling
US7623674B2 (en) * 2003-11-05 2009-11-24 Cognex Technology And Investment Corporation Method and system for enhanced portal security through stereoscopy
US7158225B2 (en) * 2004-01-23 2007-01-02 Kaiser Optical Systems Multi-channel, self-calibrating fiber-coupled raman spectrometers including diagnostic and safety features
JP2005229200A (ja) * 2004-02-10 2005-08-25 Fujitsu Ltd ディストーション補正回路
US20060067592A1 (en) * 2004-05-27 2006-03-30 Walmsley Simon R Configurable image processor
US20050270372A1 (en) * 2004-06-02 2005-12-08 Henninger Paul E Iii On-screen display and privacy masking apparatus and method
WO2006052508A2 (en) * 2004-11-03 2006-05-18 Tyzx, Inc. An integrated image processor

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See references of EP1810220A4 *

Also Published As

Publication number Publication date
US20060153450A1 (en) 2006-07-13
EP1810220A2 (en) 2007-07-25
JP2008519371A (ja) 2008-06-05
EP1810220A4 (en) 2008-04-30
WO2006052508A3 (en) 2007-04-26
US8724885B2 (en) 2014-05-13
US7664315B2 (en) 2010-02-16
US20100104175A1 (en) 2010-04-29
JP5306652B2 (ja) 2013-10-02

Similar Documents

Publication Publication Date Title
US8724885B2 (en) Integrated image processor
CN109886896B (zh) 一种蓝色车牌分割与矫正方法
CN112991369B (zh) 基于双目视觉的行驶车辆外廓尺寸检测方法
CN110322522B (zh) 一种基于目标识别区域截取的车辆颜色识别方法
US9426449B2 (en) Depth map generation from a monoscopic image based on combined depth cues
CN112201078B (zh) 一种基于图神经网络的自动泊车停车位检测方法
CN101527046B (zh) 一种运动检测方法、装置和系统
CN102883175B (zh) 深度图提取、判断视频场景切换及深度图边缘优化方法
CN108805042B (zh) 道路区域监控视频被树叶遮挡的检测方法
CN111127318A (zh) 一种机场环境下的全景图像拼接方法
CN117036641A (zh) 一种基于双目视觉的公路场景三维重建与缺陷检测方法
US20140233847A1 (en) Networked capture and 3d display of localized, segmented images
JP2008519371A5 (enrdf_load_stackoverflow)
CN110414385B (zh) 一种基于单应性变换和特征窗口的车道线检测方法及系统
CN104715239A (zh) 一种基于去雾处理和权重分块的车辆颜色识别方法
CN111739031A (zh) 一种基于深度信息的作物冠层分割方法
CN101630407B (zh) 基于两视几何和图分割的伪造区域定位方法
CN107507140A (zh) 基于特征融合的高速公路露天场景车辆阴影干扰抑制方法
CN113313968A (zh) 车位检测方法及存储介质
Xiang et al. Exemplar-based depth inpainting with arbitrary-shape patches and cross-modal matching
CN110969135B (zh) 自然场景中的车标识别方法
Shioyama et al. Measurement of the length of pedestrian crossings and detection of traffic lightsfrom image data
EP2466903B1 (en) A method and device for disparity range detection
CN112102347A (zh) 一种基于双目视觉的台阶检测和单级台阶高度估计方法
CN118015093A (zh) 一种弱约束环境下的lidar与camera联合标定方法、装置

Legal Events

Date Code Title Description
AK Designated states

Kind code of ref document: A2

Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BW BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE EG ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KM KN KP KR KZ LC LK LR LS LT LU LV LY MA MD MG MK MN MW MX MZ NA NG NI NO NZ OM PG PH PL PT RO RU SC SD SE SG SK SL SM SY TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW

AL Designated countries for regional patents

Kind code of ref document: A2

Designated state(s): GM KE LS MW MZ NA SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HU IE IS IT LT LU LV MC NL PL PT RO SE SI SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG

121 Ep: the epo has been informed by wipo that ep was designated in this application
WWE Wipo information: entry into national phase

Ref document number: 2007540360

Country of ref document: JP

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2005821343

Country of ref document: EP

WWP Wipo information: published in national office

Ref document number: 2005821343

Country of ref document: EP